SuperVision - CFlex RTP - a decision support system for locomotive allocation and wagon distribution

• Grant number: 07/51608-8
• Support Opportunities: Research Grants - Innovative Research in Small Business - PIPE
• Start date: March 01, 2008
• End date: February 28, 2010
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computer Systems
• Principal Investigator: Rodrigo Almeida Gonçalves
• Grantee: Rodrigo Almeida Gonçalves
• Company: CFLEX Computação Flexível Aplicada Ltda
• City: Campinas

Abstract

In 2004, FAPESP approved a request from CFlex to support Phase 1 of the SuperVision project (04-02901-6R). Given the socio-economic relevance of the project, CFlex decided to prioritize it as much as possible and, within Phase 1, Phase 2 and 3 were also completed. We assembled all resources for the project and with great effort we succeeded in achieving this goal. The result was the production of a decision support tool that we put on the market under the name of CFlex Trains, or Trains. CFlex Trains is a technological innovation and is bringing enormous productivity gains to railroads that have already implemented the system (e.g. 16.8% reduction of train stop time in Latin America Logistics). Currently it is implanted in 2 Brazilian railroads and is in implantation in 3 other Brazilian and 1 Argentinean ones. We are also in the process of negotiating the product with Australian, New Zealand, Mexican and South African railroads. Through this document we are requesting to FAPESP a new support to develop another technological innovation, within the same context of the SuperVision project. It is a system to support the decision-making, at the operational level, for the allocation of assets of railways, that is, locomotives and wagons. There are already some products on the market that aim to optimize the allocation of railroad assets, but none of them at the operational level, that is, none of them are part of the daily routine of asset management of the railways. Existing systems are useful in strategic and tactical medium- and long-term planning, but not at the operational level, that is, in real-time, short-term decision making. The proposed system, called CFlex RTP, is a paradigm shift in asset planning. In addition to developing new algorithms and optimization techniques, our goal includes the development of man - machine interaction mechanisms that allow the integration of the system to the operational reality of companies. (AU)